Is remote viewing real? What the 1995 review found
Nobody has settled the question, but the US government's own 1995 review came close to a verdict. The American Institutes for Research found that laboratory hits occurred more often than chance, that the design of the studies could not show why, and that in no case had remote viewing information been used to guide an intelligence operation. The program was shut down the same year.
Who reviewed it and how
What the laboratory review found
What the operational review found
Where the two reviewers disagreed
Earlier verdicts in the files
Where to read it
What the documents say, and what people claim
Who reviewed it and how
In 1995 Congress moved the Star Gate program to the CIA, and the CIA hired the American Institutes for Research (AIR), a private nonprofit, to decide whether it should continue. The report, "An Evaluation of Remote Viewing: Research and Applications," is dated September 29, 1995, and is signed by three AIR scientists, Michael Mumford, Andrew Rose and David Goslin. The CIA's copy runs 206 pages and was released in 2002.
AIR split the job in two. For the laboratory research it convened a panel built to disagree: Jessica Utts, a statistics professor at the University of California, Davis, who had published in favor of paranormal interpretations, and Ray Hyman, a psychology professor at the University of Oregon and a long-standing skeptic. For the operational side, AIR studied how intelligence customers had used the viewers' reports and interviewed those involved. AIR is a contractor, so its report is not public-domain government work; we summarize it and link the record.
What the laboratory review found
The panel agreed on one thing: in the laboratory studies, hits occurred more often than chance would predict. That is the "statistically significant effect" that supporters quote.
The panel also agreed on what the studies could not show. The same few viewers, the same judge and the same pool of target photographs were used again and again, so the effect could not be separated from the judges, the targets or the method. In the report's own words, "it remains unclear whether the existence of a paranormal phenomenon, remote viewing, has been demonstrated." The studies said nothing about where the effect came from, if it existed at all.
What the operational review found
This is the part that closed the program. AIR looked at the reports viewers had written for intelligence customers and at how those customers had used them. It found that the reports were sometimes right about broad background features of a target but wrong on specifics, inconsistent, and so vague that the reader had to interpret them. Most importantly, it found no case in which the information had ever been used to guide an intelligence operation.
On that basis AIR concluded the method could not produce intelligence good enough to act on and recommended that the program end; the CIA ended it. The CIA's own one-line summary, as the FAS history puts it, was that in no case had ESP provided data used to guide operations.
Where the two reviewers disagreed
Utts read the laboratory numbers as evidence that psychic functioning is real, and said so in her review. Hyman accepted the statistics but not the conclusion: in his view the studies had not ruled out ordinary explanations and the claim was premature. The report prints both reviews, Utts's reply, and their points of agreement and disagreement. Thirty years on, both camps still cite the same report.
Earlier verdicts in the files
The 1995 review was not the first. A 1975 analysis of Pat Price's four-day session on a Soviet site praised one sketch and still judged the experiment unsuccessful; see Did Pat Price really see a Soviet crane?. A 1995 CIA memo notes that the National Research Council reviewed an Army training program in 1988 and recommended that it be discontinued. And a 1982 Army memo on the Dozier kidnapping found the viewers' data "did not play a direct and immediate role" in his release; see Did remote viewing help find General Dozier?.
The documents say: a real statistical effect in the laboratory, no agreed explanation for it, and no operational use in twenty years of trying. The reviewers disagreed on what the effect meant, and the report says so.
People claim: that the government "proved" remote viewing and then hid it, or that the review "debunked" it completely. Neither matches the text. The report says the question of a paranormal cause is unresolved and that the method was useless for intelligence. Both halves are in the same document.
Last updated November 4, 2026.
READ THE DOCUMENTS OFFLINE, IN LUCID DREAM TRAINING & JOURNAL
All four declassified documents behind these answers are bundled, page for page and readable offline, in Lucid Dream Training & Journal, with a private dream journal that never touches a server.
Star Gate remote viewing, answered: What Project Star Gate was, what the 1995 review found, the Mars session, the Ark of the Covenant target, the CRV manual, Uri Geller at SRI and more, with CIA record IDs.
What was Project Star Gate?: Star Gate was the last name of a US remote viewing effort that ran from the 1970s to 1995. Dates, agencies, cost and the 1995 review, from the records.
Did Pat Price really see a Soviet crane?: A 1975 analysis of Pat Price's remote viewing of the Soviet site URDF-3 praised his crane sketch and still called the experiment unsuccessful. Why.
Is the Gateway Process real, and does it work?: The 1983 Army report found Gateway plausible and proposed experiments. No declassified record reports results. What the documents do and do not show.